Job Title
SoC Architect
Role Summary
Lead the architecture of next-generation, AI-focused server-class system-on-chip (SoC) products. Translate product requirements into SoC-level architectures and collaborate with subsystem architects, implementation teams, software, and third-party IP partners to bring designs from definition through productization.
Experience Level
Mid-level (Level - Mid-Career). The posting requests experienced SoC architects but does not state specific years of experience.
Responsibilities
Key responsibilities include defining and validating the SoC architecture and driving implementation trade-offs.
- Translate Product Requirements Documents (PRDs) into SoC-level architectures in collaboration with subsystem architects and cross-functional teams.
- Define SoC architecture: memory systems, interconnect fabrics, clocking, power architecture, IP requirements, and system integration.
- Define system flows for power-up, boot, and recovery.
- Own PPA feasibility and drive power, performance, and area trade-offs with implementation teams.
- Author and maintain architectural specifications and documentation from definition through productization.
- Coordinate across hardware, firmware/software, and third-party IP teams to ensure integration and delivery.
Requirements
Must-have technical skills, workplace requirements, and other conditions.
-
Must-have: Industry experience in complex SoC architecture and system-level design.
-
Must-have: Knowledge of memory systems and interconnect fabrics, including relevant industry standards.
-
Must-have: Ability to define functional and PPA requirements from use cases and make trade-off decisions.
-
Must-have: Strong technical communication, documentation, and organizational skills.
-
Must-have: Experience collaborating across hardware, software, and third-party IP teams.
-
Must-have: Programming/scripting skills (examples: Verilog, Bash, C/C++, Python).
-
Must-have: Must be authorized to work in the United States and able to work from the Mountain View, CA office Tuesday–Thursday (hybrid schedule).
-
Must-have: Ability to comply with U.S. export control requirements applicable to the role.
-
Nice-to-have: Prior experience with server-class AI accelerators or similar high-performance compute SoCs.
Education Requirements
Not specified.
About the Company
Company: MatX
Headquarters: Mountain View, California, USA
MatX specializes in creating faster chips for large language models (LLMs), focusing on innovative hardware and software solutions. The company fosters a collaborative and supportive work environment, welcoming candidates of all experience levels. Their approach prioritizes deep understanding and consideration of novel methods to drive efficiency and performance in their projects, particularly in silicon design and related engineering roles.

Date Posted: 2026-07-22